Jackrong/Llama-3.1-8B-Think-Zero-GRPO

Uploaded finetuned model

• Developed by: Jackrong
• License: apache-2.0
• Finetuned from model : unsloth/Llama-3.1-8B-Instruct
• Note: Llama‑3.1‑8B‑Think‑Zero‑GRPO is a variant of the model where it was trained solely with Group Relative Policy Optimization (GRPO), using mathematics alone, and only after a tiny amount of cold‑start data. It is an intermediate version of Llama3.1‑8B‑Thinking‑R1.